From blackjack to Wall Street
Thorp recounts inventing card-counting to beat casino blackjack, then carrying the same probabilistic edge into markets through his hedge fund. One idea, an exploitable edge, connects the casino and the exchange.
A Man for All Markets: From Las Vegas to Wall Street, How I Beat the Dealer and the Market
7 ideas
Find an edge, then size bets
Thorp pairs identifying a statistical edge with disciplined position sizing (Kelly-style) to compound gains while surviving variance. Neither the edge nor the sizing works without the other.
Markets are beatable with rigor, not tips
He argues quantitative, evidence-based methods and low costs outperform speculation and hot tips. The account is a practitioner's case against passive fatalism about market efficiency.
Kelly criterion sizes bets to maximize growth
Once you have an edge, bet a fixed fraction of current bankroll roughly equal to edge divided by the odds. This maximizes long-run compound growth while making ruin nearly impossible. Betting more than the Kelly fraction increases volatility and lowers growth, so an overbettor with a real edge can still go broke.
Card counting exploits blackjack's shifting deck
Unlike roulette, blackjack has memory: each dealt card changes the composition of the cards that remain. When the remaining deck is rich in tens and aces, the player gains a positive expectation. Tracking the ratio of high to low cards and betting big only in those moments converts a house-edge game into a player-edge game.
Markets contain exploitable, mathematically priceable mispricings
Markets are largely but not fully efficient, and a precise pricing model can reveal specific securities trading away from fair value. Thorp priced stock warrants and convertibles with a formula equivalent to Black-Scholes before its publication.
Thorp exposes Madoff's impossible trades in 1991
In 1991 a consulting client asked Thorp to review its Madoff investment, whose returns were suspiciously smooth. Thorp checked Madoff's reported option trades against public exchange records. Many trades could not have happened, some exceeding the total volume traded that day, and he advised the client to withdraw, 17 years before the fraud collapsed.